# AttTrack: Online Deep Attention Transfer for Multi-object Tracking

arXiv: https://arxiv.org/abs/2210.08648 · HTML: https://mlsanity.com/paper/2210.08648/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 43 references that are in the map.
- First posted on arXiv 16 Oct 2022.

## Builds on (most cited first, 23 of 23)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [You Only Look Once: Unified, Real-Time Object Detection](https://mlsanity.com/paper/1506.02640/) (2015)
- [Rich feature hierarchies for accurate object detection and semantic segmentation](https://mlsanity.com/paper/1311.2524/) (2013)
- [FitNets: Hints for Thin Deep Nets](https://mlsanity.com/paper/1412.6550/) (2014)
- [Pruning Filters for Efficient ConvNets](https://mlsanity.com/paper/1608.08710/) (2016)
- [Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfer](https://mlsanity.com/paper/1612.03928/) (2016)
- [XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks](https://mlsanity.com/paper/1603.05279/) (2016)
- [A Survey of Quantization Methods for Efficient Neural Network Inference](https://mlsanity.com/paper/2103.13630/) (2021)
- [FairMOT: On the Fairness of Detection and Re-Identification in Multiple Object Tracking](https://mlsanity.com/paper/2004.01888/) (2020)
- [Deep Layer Aggregation](https://mlsanity.com/paper/1707.06484/) (2017)
- [Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activations](https://mlsanity.com/paper/1609.07061/) (2016)
- [HigherHRNet: Scale-Aware Representation Learning for Bottom-Up Human Pose Estimation](https://mlsanity.com/paper/1908.10357/) (2019)
- [Deconstructing Lottery Tickets: Zeros, Signs, and the Supermask](https://mlsanity.com/paper/1905.01067/) (2019)
- [Deep Learning with Low Precision by Half-wave Gaussian Quantization](https://mlsanity.com/paper/1702.00953/) (2017)
- [Reverse Attention for Salient Object Detection](https://mlsanity.com/paper/1807.09940/) (2018)
- [Towards Universal Object Detection by Domain Attention](https://mlsanity.com/paper/1904.04402/) (2019)
- [AR-Net: Adaptive Frame Resolution for Efficient Action Recognition](https://mlsanity.com/paper/2007.15796/) (2020)
- [DeepCache: Principled Cache for Mobile Deep Vision](https://mlsanity.com/paper/1712.01670/) (2017)
- [AdaScale: Towards Real-time Video Object Detection Using Adaptive Scaling](https://mlsanity.com/paper/1902.02910/) (2019)
- [Towards Accurate Pixel-wise Object Tracking by Attention Retrieval](https://mlsanity.com/paper/2008.02745/) (2020)
- [Object detection at 200 Frames Per Second](https://mlsanity.com/paper/1805.06361/) (2018)
- [DeepScale: Online Frame Size Adaptation for Multi-object Tracking on Smart Cameras and Edge Servers](https://mlsanity.com/paper/2107.10404/) (2021)

## Built on by (most cited first, 1 of 1)

- [Edge Video Analytics: A Survey on Applications, Systems and Enabling Techniques](https://mlsanity.com/paper/2211.15751/) (2022)

## Abstract (arXiv)

Multi-object tracking (MOT) is a vital component of intelligent video analytics applications such as surveillance and autonomous driving. The time and storage complexity required to execute deep learning models for visual object tracking hinder their adoption on embedded devices with limited computing power. In this paper, we aim to accelerate MOT by transferring the knowledge from high-level features of a complex network (teacher) to a lightweight network (student) at both training and inference times. The proposed AttTrack framework has three key components: 1) cross-model feature learning to align intermediate representations from the teacher and student models, 2) interleaving the execution of the two models at inference time, and 3) incorporating the updated predictions from the teacher model as prior knowledge to assist the student model. Experiments on pedestrian tracking tasks are conducted on the MOT17 and MOT15 datasets using two different object detection backbones YOLOv5 and DLA34 show that AttTrack can significantly improve student model tracking performance while sacrificing only minor degradation of tracking speed.